Originals
The association between bronchial anthracofibrosis and pneumoconiosis: A retrospective cross-sectional study
2015 年 57 巻 2 号 p. 110-117
詳細
2015 年 57 巻 2 号 p. 110-117
Objectives: Bronchial anthracofibrosis (BAF) is associated with occupational hazardous dust exposure. The aim of the present study was to determine the prevalence of BAF and BAC without fibrosis in patients with pneumoconiosis, and to evaluate the associations between BAC/BAF and occupational dust exposure and clinical manifestations among patients with pneumoconiosis. Methods: A retrospective cross-sectional study (n=170) among individuals who were diagnosed with pneumoconiosis or suspicious pneumoconiosis and underwent bronchoscopy between January 2000 and February 2013 was performed. Multiple logistic regression analysis was performed to estimate associations. Results: In total, 153 eligible subjects were included in the study because their records contained all the required information. Of these, 81 (53%) and 63 (41%) had BAC and BAF, respectively. Occupational coal dust exposure increased the risk of BAF and BAC (odds ratio [OR]=2.980, 95% confidence interval [CI]=1.184–8.128; OR=2.840, 95% CI=1.092–7.926, respectively). Profusion category 3 pneumoconiosis also increased the risk of BAC (OR=33.887, 95% CI=5.317–394.729). Conclusions: BAF and BAC are associated with occupational exposure to coal dust. Therefore, clinicians should consider occupational history when they investigate the association between BAC/BAF and risk factors such as tuberculosis, lung cancer, and biomass fuel exposure.
(J Occup Health 2015; 57: 110–117)
Bronchial anthracosis (BAC) is defined as black tattoos on the bronchial wall surface due to the deposition of coal or other carbon particles in the bronchial wall1, 2). Bronchial anthracofibrosis (BAF) is defined as bronchial narrowing or obliteration that is associated with BAC3). Patients with BAF frequently complain that their sputum is hard to excrete physiologically due to wall narrowing4). They also have a high incidence of pneumonia (up to 40%)5). The location of pneumonic consolidation on a plain chest radiograph often corresponds to the lobes with bronchial narrowing, and this confirms the diagnosis of pneumonia6). Thus, BAF may be a major risk factor of pneumonia4, 7).
BAF itself is associated with several risk factors, namely, chronic obstructive pulmonary disease (COPD), lung cancer, and tuberculosis3, 5, 7, 8). Women in developing countries who are exposed over long periods to biomass fuel exhaust also often develop BAF3, 5). Moreover, a recent case series report from a British hospital and a letter about that report suggest that BAC and BAF are common in coal workers and other workers who are exposed to dust9, 10). However, only two studies have assessed the association of BAF with occupational dust exposure history: both of them were case reports2, 10). One retrospective study has evaluated the clinical features of BAF in 32 Korean subjects with coal worker pneumoconiosis, but only radiological features were described: the detailed occupational history of the patients was not discussed11). Thus, large-scale studies on the association between BAC/BAF and occupational dust exposures have not yet been performed.
The aim of the present study was to determine the prevalence of BAF and BAC without fibrosis in patients with pneumoconiosis and to evaluate the associations between BAC/BAF and occupational dust exposure and clinical manifestations among patients with pneumoconiosis.
This cross-sectional study was performed by retrospectively examining the hospital records of all 170 patients with pneumoconiosis or suspicious pneumoconiosis (diagnosed by using the International Labour Organization (ILO) pneumoconiosis classification system) who underwent bronchoscopy between January 1, 2000, and February 27, 2013, at the Departments of Occupational and Environmental Medicine of Yeouido St. Mary's Hospital (January 1, 2000 to April 30, 2011) and Seoul St. Mary's Hospital (since May 1, 2011). On May 1, 2011, the Department of Occupational and Environmental Medicine at Yeouido St Mary's Hospital was closed, and all patients were referred to Seoul St. Mary's Hospital. Both hospitals are teaching hospitals and specified for treating patients with pneumoconiosis. Seventeen cases were excluded because of missing data, that is, the bronchoscopy reports were incomplete and/or critical occupational information was lacking. The remaining 153 patients were included in the study. This study was approved by the Institutional Review Board of Seoul St. Mary's Hospital (ID: KC13RISI0778).
Demographic and clinicopathological characteristicsTable 1 summarizes the demographic and clinicopathological characteristics of the patients with pneumoconiosis, namely, their age, height, weight, and smoking behavior. Smoking behavior was expressed as smoking duration and smoking status (never-smoker, ex-smoker, and current smoker). Table 1 also contains the final bronchoscopy results of the patients, which were classified as no abnormality, lung cancer, tuberculosis or other (neuroma, post-tracheostomy stenosis, candida infection, bronchitis, non-tuberculous mycobacteria, fibrosis, usual interstitial pneumonia, and pneumonia). For those patients who underwent bronchoscopy more than once, only the most recent result was included.
| BAF | BAC | |||||
|---|---|---|---|---|---|---|
| No | Yes | p-value | No | Yes | p-value | |
| Age (year) | 66.1 (8.6) | 67.5 (9.2) | 0.338 | 65.9 (8.4) | 67.4 (9.2) | 0.300 |
| Smoking duration (pack-years) | 21.0 (19.7) | 20.2 (21.2) | 0.824 | 21.9 (18.9) | 19.5 (21.5) | 0.470 |
| Smoking status* | ||||||
| Never-smoker | 25 (28.1) | 25 (39.7) | 0.315 | 18 (25.4) | 32 (39.5) | 0.175 |
| Ex-smoker | 58 (65.2) | 35 (55.6) | 48 (67.6) | 45 (55.6) | ||
| Current smoker | 6 (6.7) | 3 (4.7) | 5 (7.0) | 4 (4.9) | ||
| Final diagnosis of bronchoscopy† | ||||||
| No abnormality | 46 (51.1) | 44 (74.6) | 0.005 | 33 (44.5) | 58 (75.3) | <0.001 |
| Lung cancer | 29 (32.2) | 5 (8.4) | 26 (36.1) | 8 (10.4) | ||
| Tuberculosis | 8 (8.9) | 7 (11.9) | 7 (9.7) | 8 (10.4) | ||
| Etc‡ | 7 (7.8) | 3 (5.1) | 7 (9.7) | 3 (3.9) | ||
| Occupational exposure duration (year) | 20.2 (8.4) | 18.9 (8.9) | 0.370 | 20.7 (8.3) | 18.7 (8.8) | 0.161 |
| Silica exposure duration (year) | 7.2 (10.7) | 5.1 (8.6) | 0.186 | 7.0 (11.0) | 5.7 (8.9) | 0.424 |
| Coal exposure history* | ||||||
| No | 31 (34.8) | 16 (25.4) | 0.215 | 25 (35.2) | 22 (27.2) | 0.284 |
| Yes | 58 (65.2) | 47 (74.6) | 46 (64.8) | 59 (72.8) | ||
| Rock drilling history* | ||||||
| No | 59 (66.3) | 43 (68.3) | 0.800 | 50 (70.4) | 52 (64.2) | 0.284 |
| Yes | 30 (33.7) | 20 (31.7) | 21 (29.6) | 29 (35.8) | ||
| Plain chest radiologic findings of pneumoconiosis** | ||||||
| Profusion | ||||||
| 0 | 5 (6.5) | 1 (1.7) | 0.029 | 4 (6.5) | 2 (2.7) | 0.002 |
| 1 | 36 (46.7) | 18 (31.0) | 32 (51.6) | 22 (30.1) | ||
| 2 | 30 (39.0) | 26 (44.9) | 24 (38.7) | 32 (43.8) | ||
| 3 | 6 (7.8) | 13 (22.4) | 2 (3.2) | 17 (23.3) | ||
| Opacity | ||||||
| No PMF | 24 (32.0) | 12 (20.7) | 0.129 | 19 (31.7) | 17 (23.3) | 0.572 |
| A | 33 (44.0) | 22 (37.9) | 25 (41.7) | 30 (41.1) | ||
| B | 16 (21.3) | 19 (32.8) | 14 (23.3) | 21 (28.8) | ||
| C | 2 (2.7) | 5 (8.6) | 2 (3.3) | 5 (6.8) | ||
| Total | 90 (100.0) | 63 (100.0) | 72 (100.0) | 81 (100.0) | ||
BAF=bronchial anthracofibrosis; BAC=bronchial anthracosis; PMF=progressive massive fibrosis, any opacity greater than 1 cm in diameter; A=for one or more large opacities whose combined longest dimension does not exceed about 50 mm; B=for one or more larege opacities whose combined longest dimension exceed 50 mm but does not exceed the equivalent area of the right upper lung; C=for one or more large opacities whose combined longest dimension exceed the equivalent area of the right upper lung zone. For continuous variables, the Students t test was used for the analysis.
We determined the operational definition of BAF as fibrosis of the bronchus combined with black pigment deposition resulting in one or more narrowed or stenosed bronchus orifices. BAC was defined as black pigment deposition due to exogenous dust (coal, silica, carbon particle, etc) without fibrosis. The clinician reports of the most recent bronchoscopy were reviewed for mention of BAF and BAC findings. However, the primary purpose of bronchoscopy was for purposes other than evaluating BAF or BAC, and thus BAF/BAC was sometimes not mentioned. In such cases, the captured images were rechecked by Drs. JP Myong and YM Cho. The kappa of interobserver agreement for BAC was 1.000, and that for BAF was 0.819 (the 95% confidence intervals (95% CI): 0.650–0.987). To approach the robust result, only if the opinion of two doctors was consistent was the captured image accepted as having BAF or BAC.
Radiologic findings of pneumoconiosis and occupational evaluationThe plain chest radiological findings of pneumoconiosis are shown in Table 1. They are expressed in terms of profusion and large opacity. Profusion refers to the concentration of small opacities (<1 cm in diameter), which is classified by the ILO classification system on a 4-point category scale (0–3). The profusion scores (0–3) were reclassified based on a 3-point scale, namely 0/0∼1, 2 and 3 for logistic regression. Category 0 indicates that small opacities are absent, while category 3 indicates the most severe profusion. Large opacities are defined as opacities that exceed 1 cm in diameter. They are classified as categories A (one or more large opacities whose combined longest dimension does not exceed 50 mm), B (one or more large opacities whose combined longest dimension exceeds 50 mm but does not exceed the equivalent area of the right upper lung zone) and C (one or more large opacities whose combined longest dimension exceeds the equivalent area of the right upper lung zone). The occupational history for all patients was determined by examining their medical records (which stated whether the patients had been coal workers) and by interviewing them in the ward or outpatient clinic for exposure to silica or coal. Whether the patients had a rock drilling history was also determined. The silica and total occupational exposure durations were calculated.
Statistical analysisThe pneumoconiosis patient group was divided according to whether the patients had BAF, and the two groups were compared in terms of demographic, clinicopathological, occupational exposure and radiologic variables by using the Student's t-test, the Chi-square test and logistic regression analysis. Similarly, the subjects were divided according to whether they had BAC, and the same analysis was performed. In multiple logistic regression analysis, age, smoking history, coal exposure duration, profusion category, large opacity category and the diagnosis of the most recent bronchoscopy were included. Statistical Analysis System (SAS) version 9.2 (SAS Institute, Cary, NC, USA) was used to analyze the data. p-values of <0.05 were considered to indicate statistical significance.
Eighty-one and 63 study subjects were diagnosed as having BAC and BAF, respectively. Comparison of the patients with and without BAF revealed that the patients with BAF did not differ significantly from the patients without BAF in terms of age, height, weight, smoking duration, and smoking status. The two groups did differ significantly in terms of final bronchoscopy diagnosis: lung cancer seemed to be rarer in the patients with BAF than the patients who did not have BAF (p<0.001). The patients without clinical abnormalities had higher frequencies of BAF (74.6%) or BAC (75.3%) than the patients with lung cancer, tuberculosis, or other diseases, while the patients with lung cancer had frequencies of BAF and BAC of 8.4 and 10.4%, respectively.
As also shown in Table 1, 105 of the 153 patients had a history of coal exposure. Forty-seven (74.6%) and 59 (72.8%) had BAF and BAC, respectively. In addition, 50 of the 153 patients had a history of rock drilling (31.7% for BAF (n=20) and 35.8% for BAC (n=29)). Comparison of the patients with and without BAF revealed that the patients with BAF did not differ from the patients without BAF in terms of occupational exposure duration, silica exposure duration, coal exposure history or large opacity category. However, the patients with BAF were more likely to exhibit category 3 profusion than the patients without BAF (p=0.029). This was also true for BAC (p=0.002).
Table 2 indicates the unadjusted associations of BAF/BAC with age, smoking status, coal exposure, profusion, large opacity category, and diagnosis at the most recent bronchoscopy. BAF was significantly associated with category 3 profusion (the reference was a composite of categories 0, 0/1, and 1): the odds ratio (OR) was 4.675, and the 95% confidence interval (95% CI) was 1.596–15.115. BAF was also significantly associated with opacity categories B and C (the reference was category A): the ORs (95% CI) were 2.375 (1.920–6.344) and 4.998 (1.930–38.666), respectively. BAF was also negatively associated with a risk of lung cancer: the OR (95% CI) was 0.152 (0.049–0.385). Similarly, BAC was associated with category 3 profusion, large opacity category A and lung cancer: the ORs (95% CI) were 12.750 (3.262–85.149), 1.341 (1.578–3.138) and 0.170 (0.065–0.403), respectively.
| BAF | BAC | |||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Age (per year) | 1.018 | (0.982–1.058) | 1.019 | (0.983–1.058) |
| Smoking status | ||||
| Never-smoker | 1.000 | 1.000 | ||
| Ex-smoker | 0.603 | (0.300–1.209) | 0.527 | (0.257–1.060) |
| Current smoker | 0.500 | (0.090–2.119) | 0.450 | (0.100–1.905) |
| Coal exposure | ||||
| No | 1.000 | 1.000 | ||
| Yes | 1.570 | (0.775–3.265) | 1.458 | (0.73–2.925) |
| Profusion | ||||
| 0 or 0/1 or 1 | 1.000 | 1.000 | ||
| 2 | 1.870 | (0.883–4.025) | 2.000 | (0.960–4.232) |
| 3 | 4.675 | (1.596–15.115) | 12.750 | (3.262–85.149) |
| Opacity | ||||
| No PMF | 1.000 | 1.000 | ||
| A | 1.333 | (0.559–3.268) | 1.341 | (1.578–3.138) |
| B | 2.375 | (1.920–6.344) | 1.676 | (0.658–4.358) |
| C | 4.998 | (1.930–38.666) | 2.793 | (0.526–21.334) |
| Final diagnosis of bronchoscopy† | ||||
| No abnormality | 1.000 | 1.000 | ||
| Lung cancer | 0.180 | (0.057–0.472) | 0.170 | (0.065–0.403) |
| Tuberculosis | 0.915 | (0.298–2.756) | 0.631 | (0.208–1.949) |
| Etc‡ | 0.448 | (0.092–1.723) | 0.236 | (0.048–0.913) |
BAF=bronchial anthracofibrosis; BAC=bronchial anthracosis.
Table 3 indicates the multiple logistic regression associations of BAF/BAC after adjustments for age, smoking status, coal exposure history, profusion category, large opacity category, and the most recent bronchoscopy diagnosis. The risk of BAF was significantly higher when there was a coal exposure history (OR=2.980, 95% CI=1.184–8.129) or category 3 profusion (OR=4.370, 95% CI=1.138–18.933) and significantly lower if the patient was diagnosed with lung cancer on the most recent bronchoscopy (OR=0.140, 95% CI=0.036–0.445). Similarly, the risk of BAC was significantly higher if the patient had a coal exposure history (OR=2.840, 95% CI=1.092–7.926) or category 2 (OR=2.650, 95% CI=1.079–6.867) or 3 profusion (OR=33.877, 95% CI=5.317–394.729) and significantly lower if the patient was diagnosed with lung cancer on the most recent bronchoscopy (OR=0.111, 95% CI=0.032–0.330).
| BAF | BAC | |||
|---|---|---|---|---|
| OR† | 95% CI† | OR† | 95% CI† | |
| Coal exposure history | ||||
| No | 1.000 | 1.00 | ||
| Yes | 2.980 | (1.184–8.128) | 2.840 | (1.092–7.926) |
| Profusion | ||||
| No pneumoconiosis or 1 | 1.000 | 1.000 | ||
| 2 | 2.031 | (0.841–5.077) | 2.650 | (1.079–6.867) |
| 3 | 4.370 | (1.138–18.933) | 33.877 | (5.317–394.729) |
| Opacity | ||||
| No PMF | 1.000 | 1.000 | ||
| A | 1.631 | (0.584–4.709) | 1.705 | (0.594–5.040) |
| B | 2.040 | (0.669–6.440) | 0.824 | (0.215–2.647) |
| C | 2.707 | (0.352–26.711) | 0.634 | (0.043–9.021) |
| Final diagnosis | ||||
| No significant abnormality | 1.000 | 1.000 | ||
| Lung cancer | 0.140 | (0.036–0.445) | 0.111 | (0.032–0.330) |
| Tuberculosis | 0.588 | (0.133–2.435) | 0.434 | (0.093–1.939) |
| Etc‡ | 0.273 | (0.043–1.315) | 0.102 | (0.011–0.599) |
BAF=bronchial anthracofibrosis; BAC=bronchial anthracosis; PMF=progressive massive fibrosis, any opacity greater than 1 cm in diameter; A=for one or more large opacities whose combined longest dimension does not exceed about 50 mm; B=for one or more larege opacities whose combined longest dimension exceed 50 mm but does not exceed the equivalent area of the right upper lung; C=one or more large opacities whose combined longest dimension exceed the equivalent area of the right upper lung zone.
(A) Bronchoscopic examination showed anthracosis in the left upper bronchus; 79-year-old man, 9 years of exposure to coal dust in a coal mine and manufacturing of briquette (B) Bronchoscopic examination revealed anthracotic fibrosis in the left upper bronchus; 70-year-old man, 18 years of exposure to coal dust in a coal mine (C) Axial CT image with lung window settings demonstrating anthracofibrosis with distal collapse; 70-year-old man, 18 years of exposure to coal dust in a coal mine.
This retrospective cross-sectional study showed that BAF and BAC were highly prevalent in the patient with pneumoconiosis, as indicated by the bronchoscopic findings. BAF was also significantly associated with a history of coal exposure and category 3 profusion. However, BAF was not associated with pulmonary tuberculosis, which contradicts the findings of previous reports3, 8, 12–14).
BAC is characterized by the deposition and accumulation of extrinsic particles such as coal dust and biomass fuel smoke in the lymph nodes as anthracite particles; it is also characterized by the tattooing of the bronchial wall surface by various particles, namely, anthracite and biomass fuel smoke15). The mechanism by which BAF by which it develops is not yet clear, although occupational coal exposure is believed to be a major source of the carbon particles10, 11). One possible mechanism is the deposition of coal dust onto the bronchus irritating the bronchus mucosa, causing the bronchus to thicken16). Another possibility is that coal and silica dust can reach the proximal airway, where they activate macrophages, which in turn induce inflammation in the interstitial space17). Laboratory studies on the mechanism by which dust exposure promotes BAF are warranted.
A previous study concerning BAF and BAC in Korean patients with coal worker pneumoconiosis revealed that they had high incidences of BAC (81%) and BAF (28%)11), which is consistent with our study (41% for BAF, 53% for BAC). However, another study showed that the profusion category did not seem to be markedly associated with the frequency of BAC or BAF11). This was not consistent with our results, but this discrepancy may reflect the relatively small sample size of the previous study by Kim et al. (n=32). Indeed, to compensate for the relatively small number of subjects with pneumoconiosis in the present study, we had to use a 12-year study period that started in 2000.
In our study, we reclassified the profusion scores (0–3) based on a 3-point scale, namely, 0/0–1, 2 and 3. It is possible that the results were biased as a result of this unusual recoding of profusion. As the category 0 patients were very small in number (n=6), the comparison among the different categories of profusion may be diluted, as the reference category was basically category 1. Therefore, we preformed an additional analysis for Table 3 in which we used four profusion categories (0,1, 2 and 3), and the resulting ORs for categories 2 and 3 were similar to those in the previous analysis. If there had been a sufficient number of subjects in the ORs for the category 0, the ORs for the other categories may have been greater. This requires further research.
Workplace exposure was not considered in a previous study11). In Korea, clinicians can diagnose CWP among patients with radiologiclly confirmed pneumoconiosis who had worked in a coal mine. However, in general, the coal workers are exposed to different dusts according to their jobs. For example, coal miners are exposed to coal dust, on the other hand, rock drillers and excavators in coal mines are exposed to silica dust. This might be a significant determining factor for coal tattoos caused by coal dust. However, the previous studies on BAC/BAF did not analyzed the role of such occupations in the risk of BAC/BAF. The detailed occupational histories of the subjects in our study supports the notion that coal exposure history is associated with an increased risk of BAF or BAC.
The present study did not find a significant association between BAC/BAF and tuberculosis. By contrast, another study noted that patients with tuberculosis frequently have BAF (46.2%) and that BAF is a significant risk factor of tuberculosis18). This discrepancy may be related to the fact that our patient population consisted of patients with pneumoconiosis. It may also reflect the fact that the previous study did not investigate the contribution of occupation to the association of BAC/BAF with tuberculosis, even though several studies have indicated the importance of occupational exposure to coal in BAF9, 10). Therefore, future studies on the relationship between BAF/BAC and various risk factors should also include occupation as a variable.
The present study showed that patients with BAF or BAC had a significantly lower risk of lung cancer. Other studies have observed the opposite relationship7, 19). There are possible explanations for a lower prevalence of BAF/BAC with lung cancer. There are several possible explanations for this. Our patients had a great diversity of bronchoscopy reasons. About 45% of the bronchoscopies were done for other reasons besides ruling out lung cancer. (This is not explained in the included tables.) For pathologic diagnosis of the lung cancer, transbronchial lung biopsy and CT-guided percutaneous needle biopsy are major tools. In the present study, subjects with peripheral lesions who underwent CT-guided needle biopsy were not be included therefore, selection bias could not be avoided. Careful consideration should be given to this when interpreting results. Further prospective study may be required to explain causation between lung cancer and anthracofibrosis.
BAF and BAC are regarded as indicators of exposure to polycyclic aromatic hydrocarbons and other hydrocarbons from diesel exhaust particles, biomass smoke, and other sources20), which are considered to be lung carcinogens by the International Agency for Research on Cancer. However, a study among Koreans with lung cancer suggested that BAF occurs less frequently in nonmalignant lymph nodes than in malignant lymph nodes21). In addition, the finding of multifocal bronchial narrowing was a critical point for differentiation from lung cancer22). This was concordant with our study results showing that BAF and BAC were more prevalent among coal-exposed subjects than other workers. Almost 41% and 53% of participants had multifocal BAF and BAC, respectively, according to bronchoscopy. As BAF or BAC was frequently found in subjects with massive coal exposures, it might be difficult to verify whether BAF or BAC represents necessary or sufficient conditions for lung cancer. Prospective studies that assess the relationship between lung cancer and BAF/BAC while including coal exposure as a variable are warranted.
The presence of BAC and/or BAF was associated with highly profused pneumoconiosis in the present study. Bronchoscopy cannot be used to observe terminal bronchioles, which is where exogenous dust accumulates. Sedimentation, impaction, Brownian diffusion, etc., are possible mechanisms explaining exogenous dust accumulation in the lung23). According to previous research, sound and deep breathing result in uniform dust deposition at the terminal bronchioles via sedimentation23, 24). However, rapid and shallow breathing due to vigorous work are able to enhance particle deposition in the central bronchus to bronchiole23, 24). As coal mine workers usually perform high-intensity labor in a hot and humid underground environment, they tend to breathe rapidly and shallowly while working. Inappropriate mask usage, penetration through a mask filter and leakage might also contribute to inhalation of massive amounts of respirable dusts. All of these may attribute to the impaction of coal workers' lung. Therefore, BAC and/or BAF may be observed by a bronchoscopic approach in pneumoconiosis with high-grade profusion. Further prospective evaluation will be needed.
The present study had some limitations. First, since it had a cross-sectional design, causality could not be determined. Second, there may have been some information bias, as the study relied on hospital records. Moreover, the duration of coal-related work was self-reported. However, all patients were also interviewed about their occupational history at the ward or in an outpatient clinic. In addition, to ensure accurate identification of the patients with BAC/BAF, all bronchoscopic findings in the picture archiving and communication system records were assessed by two respiratory physicians. Moreover, to limit information bias, we excluded all subjects with incomplete information. Third, the study population was a highly selected population because only patients with pneumoconiosis who underwent bronchoscopy were examined. The invasiveness of the bronchoscopy procedure and the potential hazards of CT mean that subjects who did not have occupational exposure were likely to be absent. However, to mitigate this, we sought to maximize our sample size by using a long study period (12 years). Longitudinal studies that avoid these limitations are warranted. Fourth, it was not possible to perform gender analyses: only seven of the enrolled subjects were women. This reflects the reluctance in Korea to allow women to work in coal mines; women are generally employed by the coal industry to classify coal or to remove the rocks from coal piles.
Despite these limitations, our study also has several important strengths. First, it had a relatively large sample size (n=153), and 81 of these patients had BAC. This is relatively larger than the sample sizes of other studies on BAF/BAC3, 11, 25). Second, all the subjects underwent bronchoscopy, and detailed information about their occupational exposure to coal dust was obtained.
In conclusion, BAC and BAF were associated with occupation exposure to coal dust. Clinicians should consider occupational history when they assess the association between BAC/BAF and possible risk factors such as tuberculosis, lung cancer, and biomass fuel exposure.
Acknowledgments: The authors wish to acknowledge the financial support of the Catholic Medical Center Research Foundation made in the program year of 2014.
Conflict of interest: The authors declare that they have no competing interests.
